Siddhartha Reddy

Software Engineer, likes to build products. Founding team & now CTO @udaandotcom. Ex: Flipkart, and a failed startup. An introvert who doesn't like to be alone.

[Siddhartha Reddy @sids]

@simpsoka On mobile, if I want to open a new project which was not already open, I have to type the entire path. It would be nice to have a directory listing to pick from.

[Deep Ganatra @DeepXP]

A friend’s wife need A negative blood in Malad. Any leads? Please spread the word.

[Siddhartha Reddy @sids]

@lucasmeijer CC adds CLAUDE.md files from sub-directories to context the first time the agent reads a file from that path. There is a pi extension to do something similar too.

[Siddhartha Reddy @sids]

@mattpocockuk I advise everyone to disable postinstall scripts because they create a security vulnerability (in case of compromised packages).

[Siddhartha Reddy @sids]

So good. “Get feedback early” feels obvious, but it’s a totally different thing to understand it viscerally.

I thought we do this as a habit at udaan (not just engg, even biz folks). But as I read this, I felt that there is so much more we could do.

[Thorsten Ball @thorstenball]

An internal Slack post I shared with the Amp team a few weeks ago:

Building software is learning.

https://t.co/gasUel9aAv

[Thorsten Ball @thorstenball]

An internal Slack post I shared with the Amp team a few weeks ago:

Building software is learning.

https://t.co/gasUel9aAv

[Siddhartha Reddy @sids]

@vedang One trick is to use a language you are not super-familiar with, so the code an agent generates is better than the one you would write ;)

[Siddhartha Reddy @sids]

@Dimillian Yes!

[Siddhartha Reddy @sids]

Never mind, just saw this. Thanks!

[Thomas Ricouard @Dimillian]

FYI you can do this

[Siddhartha Reddy @sids]

@Dimillian Love the Codex experience in the ChatGPT iOS app. Feature request: need a way to directly open the Codex section, perhaps via a shortcut or setting for the Action Button.

[Siddhartha Reddy @sids]

@Dimillian @awwstn It doesn’t do that for me; even if I’m on a particular chat and open the app after a few minutes it’s back on the home screen with new chat.

[Siddhartha Reddy @sids]

@badlogicgames @alert090579992 @OpenAI @thsottiaux did say recently that this will change “soon”

[Tibo @thsottiaux]

@RhysSullivan Yes, but soon no

[dex @dexhorthy]

using AI for coding is a deeply technical engineering craft

most people don't approach it as so, and don't get the results we associate with high craft

but the ones who do have been sprinting ahead

more tokens wont save you, more thinking + skill + llm intuition will

have been saying this for almost 9 months now

[dax @thdxr]

i have seen enough proof now that using a coding agent is a deep skill

it's confusing because the people you see heavily using them produce horrible results

but that's because it's a skill! you can get better and the ceiling seems pretty high - this is very exciting to me

[Mitchell Hashimoto @mitchellh]

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.

As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!

I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.

It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?

88ms => 1.5ms
150K allocs => ~500 allocs

Incredible right? Nope.

My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.

This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.

The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.

Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

[Siddhartha Reddy @sids]

@balamuralimenon That’s well and good as long it doesn’t turn into slop. Claude Code also went through this several months ago and ended up in a very bad place. So far, the Codex team is doing a pretty good job though.

[Mario Zechner @badlogicgames]

recommended reading. because today the "model wellbeing" shit came up, and i reduced it to "thems just matmuls".

that's obv. an oversimplification to get a point across. personally think sentience is likely substrate independent. our current models' substrate is just insufficoent.

a model on its own is not concious/sentient under popular theories/frameworks of conciousness. not because its a big matmul machine. but because it lacks things like continuity/"state", self-maintenance, believe consolidation, embodiement, grounding, feedback loops, etc. pp.

we can close some of these gaps. e.g. a model+transcript, into which it can persist current believe, and which establishes a feedback loop does tick some (insufficient) boxes on the way to machine sentience. give the model sensors and you tick more boxes, etc.

but there are a ton more boxes that need to be ticked, before we can start worrying about "model wellbeing".

https://t.co/t0PzpJoHrx

[Siddhartha Reddy @sids]

I wholeheartedly agree with Armin here.

[Armin Ronacher ⇌ @mitsuhiko]

More musings after some people got upset about the word clanker. https://t.co/gXPC6iRP0g

[Siddhartha Reddy @sids]

@mataal @_mekin @googlemaps I did this mistake myself 🤷🏽‍♂️

[Siddhartha Reddy @sids]

@emollick Meanwhile, these tools are becoming base expectations of devs in the orgs that have adopted the, irrespective of if there is any measurable impact or not.

Recently, our Codex workspace was suspended for a couple of hours and most devs freaked out!